The anti-CD80 primatized monoclonal antibody, galiximab, is well-tolerated but has limited activity in relapsed Hodgkin lymphoma: Cancer and Leukemia Group B 50602 (Alliance)

Abstract

Relapsed Hodgkin lymphoma remains a clinical challenge, with few non-cytotoxic treatment options. CD80 is a surface antigen that normally functions as a co-stimulatory molecule but is aberrantly and uniformly expressed on Reed–Sternberg cells. Galiximab is a primatized monoclonal antibody against CD80, with a favorable toxicity profile demonstrated in other lymphomas. Cancer and Leukemia Group B (CALGB) 50602 (Alliance) tested single-agent galiximab in a highly refractory group of patients with Hodgkin lymphoma (median 3 prior regimens, 83% failing after prior stem cell transplant) to determine the efficacy. The overall response rate was 10.3% and the median progression-free survival was 1.6 months. Galiximab was well-tolerated, with minimal grade 3 or 4 toxicities. Despite this preclinical rationale, single-agent galiximab had limited activity in heavily pretreated Hodgkin lymphoma.

Introduction

Despite important strides in the treatment of Hodgkin lymphoma (HL), approximately 30% of patients with advanced disease will relapse. High-dose chemotherapy with autologous stem cell rescue can salvage approximately half of relapsed patients. The management of patients relapsing despite autologous stem cell transplant or those who are not candidates for transplant strategies is limited by poor tolerance of additional cytotoxic agents, and the median survival is typically only 2–3 years. The development of targeted and less toxic therapy is greatly needed, but is hampered by the relative rarity of the Reed–Sternberg (RS) cell within its inflammatory milieu, and the relatively recent identification of RS as being of B-cell origin [1].

CD80 is a transmembrane protein constitutively expressed on a variety of B-cell lymphomas, although the density differs by histologic subtype [2]. Several groups have demonstrated the expression of CD80 on RS cells [3–6]. CD80 is an immune co-stimulatory molecule normally present on activated B-cells, antigen-presenting cells (APCs) and T-cells [7]. Several groups showed strong and extensive CD80 expression in patient HL samples, including nodular lymphocyte predominant HL [3,5,6]. In addition, a monoclonal antibody–immunotoxin conjugate against CD80 led to effective cell kill with a decrease in clonogenic potential in Raji and several Reed –Sternberg cell lines (HDLM2, KM/H2 and L428) [6].

Galiximab (IDEC-114) is a primatized immunoglobulin G1 (IgG1) lambda monoclonal antibody directed against the CD80 antigen [2]. Its variable regions are primatized (cynomologous monkeys), and the constant regions are human. Structurally, it is indistinguishable from human antibodies and has high affinity binding to CD80 [2]. Galiximab effectively blocks CD80–CD28 interactions on T-lymphocytes, but has no significant effect on CD80–CD152 (CTLA-4) interactions [2]. CD80–CD152 interactions usually lead to down-regulation of T-cell activity, and this should therefore remain intact during galiximab therapy. Galiximab appears to work primarily via cross-linking of CD80 molecules and induction of antibody-dependent cellular cytotoxicity (ADCC), but also may inhibit cellular proliferation and up-regulate apoptotic proteins [8,9].

Galiximab has been primarily evaluated in follicular lymphoma. A phase I/II study reported a single-agent response rate of 11% in follicular lymphoma, noting that most responses were delayed and occurred more than 3 months following treatment initiation [9]. When combined with rituximab, the response rate is significantly higher in patients with both relapsed and treatment-naive follicular lymphoma, and is approximately 70% in both populations [10,11]. Aggregate safety data show that the major adverse events are infusion-related, similar to other monoclonal antibodies, and hematologic toxicity is minimal. Galiximab, with a favorable toxicity profile, could easily be incorporated into cytotoxic regimens if single-agent activity could be demonstrated. With this rationale, Cancer and Leukemia Group B (CALGB) 50602 (Alliance) was designed to establish the safety and efficacy of single-agent galiximab in patients with relapsed HL.

Patients and methods

Study design

CALGB 50602 was an open-label phase II cooperative group study of single-agent galiximab in patients with relapsed/refractory HL. Institutional review boards at participating institutions approved the study, and written informed consent was required and obtained from all patients.

Patient selection

Eligibility criteria included histologically confirmed nodular lymphocyte predominant or classical HL that was recurrent or refractory after at least two prior standard chemotherapy regimens. Patients must have been ineligible for a stem cell transplant at the time of study entry. Previous autologous and/or allogeneic stem cell transplant was allowed. Other eligibility criteria included age ≥ 18 years, Eastern Cooperative Oncology Group (ECOG) performance status ≤ 2, absolute neutrophil count ≥500/μL, platelet count ≥50 000/μL, total bilirubin ≤2.0 mg/dL (absent history of Gilbert’s syndrome), aspartate transaminase (AST) ≤ 2.5 × upper limit of normal (ULN) and creatinine ≤ 2.0 mg/dL. Pregnant and lactating females, patients with known central nervous system (CNS) involvement and those with known human immunodeficiency virus (HIV) infection were excluded.

Treatment plan

Galiximab 500 mg/m² was administered via intravenous infusion over 60 min weekly for four consecutive weeks (induction) followed by once monthly (extended induction) doses until disease progression. All patients were premedicated with diphenhydramine 50 mg IV or PO (or equivalent) and acetaminophen 650 mg PO. Treatment was intended to be delivered for a minimum of 8 weeks, and responding patients (complete/partial response or stable disease) could continue on the extended induction treatment unless there was evidence of disease progression and/or toxicity. Patients experiencing infusion reactions could have the infusion rate slowed. There were no dose modifications for galiximab, and drug was discontinued for life-threatening toxicity. Routine supportive care interventions such as erythropoietin, blood transfusions and hematopoietic colony stimulating factors for treatment of cytopenias were permitted. Concomitant steroids were prohibited except for management of severe infusion reactions.

Correlative studies

All patients were requested, but were not required, to participate in a secondary endpoint study to determine whether early fluorodeoxyglucose-positron emission tomgraphy (FDG-PET) imaging could predict response to galiximab. FDG-PET images were collected at baseline and at the time of first restaging (just prior to week 8 treatment). There were strict guidelines for PET acquisition, in accordance with the International Harmonization Criteria [12].

Response and toxicity assessment

Response assessments were performed at week 8, week 16 and then every 12 weeks. After 2 years, restaging for all patients who had not yet progressed was performed every 6 months. Response assessment was as per the revised response criteria for malignant lymphoma [13].

Study endpoints and statistical analysis

The primary endpoint of this open-label, multicenter phase II trial was complete response (CR) rate and overall response rate (ORR), defined as partial response (PR) plus CR. Based on response rates of other single agents in this population prior to the advent of brentuximab vedotin, an overall response rate of 15% or lower would be considered of low clinical interest, whereas a response rate of 30% or higher would be of interest to support further study. The two-stage admissible design [14] required four responders from the first 21 patients in order to continue to accrue an additional 28 eligible and evaluable patients, and accept the study regimen when 12 or more responders were observed from the cumulative 49 patients; this design allowed a one-sided alpha level of 5% and a power of 80%.

Secondary endpoints included duration of response, progression-free (PFS) and overall survival (OS), safety and tolerability, and to determine whether FDG-PET was associated with outcome. The PET images were collected at baseline (prior to study treatment initiation) and at week 8 of treatment. The highest activity concentration in a given disease site (maximum standardized uptake value, SUV_max) was recorded. For further analysis, we generated one composite SUV_max number per patient, at both baseline and first from up to five disease sites restaging, by adding the SUV_max (or fewer, if fewer nodal regions were involved) at each time point. The composite SUV_max at post-therapy and the ratio between the composite SUV_max at post-therapy and the composite SUV_max baseline would be used as the observation posite SUV_max for each lesion.

As part of the quality assurance program of the CALGB, members of the Audit Committee visit all participating institutions at least once every 3 years to review source documents. The auditors verify compliance with federal regulations and protocol requirements, including those pertaining to eligibility, treatment, adverse events, tumor response and outcome in a sample of protocols at each institution. Such on-site review of medical records was performed for a subgroup of two of 30 patients (6%) under this study.

Data collection and statistical analyses were conducted by the Alliance Statistics and Data Center. Data quality was ensured by review of data by the Alliance Statistics and Data Center and by the study chairperson following Alliance policies. All analyses were based on the study database frozen on 16 April 2012.

Results

Patient characteristics

Between June 2008 and January 2009, 30 patients with classical HL were accrued (Table I). There were no patients with nodular lymphocyte predominant HL. One patient never received treatment and was excluded from analysis. The median age was 36 years (range, 22–70). Patients were heavily pretreated with a median of 3 prior regimens; 72% of patients had received prior radiation therapy. Twenty-four patients (83%) had failed a prior stem cell transplant (21 autologous stem cell transplant; three allogeneic stem cell transplant). Among six patients without a prior stem cell transplant, three had refractory disease.

Table I.

Patient characteristics.

Variable	Result
Number of patients registered	30
Number of patients in analysis	29
Median age	36 years (range, 22–70)
Gender	15 male, 15 female
NLPHL	0
Median prior treatments	3 (range, 1– 7)
Prior RT	21 (72.4%)
Prior stem cell transplant	24 (82.8%)

NLPHL, nodular lymphocyte predominant Hodgkin lymphoma; RT, radiotherapy.

Response and outcome

There was one complete and one partial response among the first 21 patients treated. As the study approached the 21 patients required for the stage 1 analyses, 11 patients were registered within a 3-week period, seven of these after the pending closure notice was distributed, resulting in an unanticipated over-accrual to the study. One of nine patients accrued beyond the 21 needed for the stage 1 analysis never started treatment, and one of the nine achieved a PR.

Among the 29 treated patients there were two PRs and one CR (ORR 10.3%). (Table II; Figure 1, missing data on 3 patients). The complete responder had a PR at week 8, a CR 2 months later, but then progressed at 5 months. A second patient achieved a PR at 3 months, but progressed 3 months later (month 6). The third responding patient, also with a PR, achieved stable disease at first restaging, a PR at month 9, but then progressed. Nine patients achieved stable disease, of whom eight have progressed. One patient with stable disease received 33 cycles until the drug supply was depleted; follow-up 4 months after completing therapy showed continued stable disease for this patient.

Table II.

Response to single-agent galiximab in relapsed/refractory Hodgkin lymphoma.

Outcome	Patients, n (%)
ORR	3 (10.3%)
Complete response	1 (3.5%)
Partial response	2 (3.5%)
Stable disease	9 (31%)
Progressive disease
PD after initial SD	7 (24%)
Immediate PD	17 (58.6%)
Median TTP	1.6 months
6-Month PFS	0.21 (0.08, 0.37)
6-Month OS	0.93 (0.74, 0.98)
Median follow-up	13.6 months (range, 1.8–18.4)
Patients alive/dead	18/11

ORR, overall response rate; PD, progressive disease; SD, stable disease; TTP, time to progression; PFS, progression-free survival; OS, overall survival.

Figure 1.

Waterfall plot of percent change in tumor dimensions at the time of first restaging (8 weeks) in patients with relapsed Hodgkin lymphoma receiving single-agent galiximab, color-coded by best response. One patient with stable disease achieved a partial response at month 9. One patient with a PR at first restaging became a complete responder at month 3.

The median follow-up time in the 10 patients still alive is 34.2 months (range, 7.8–37.7 months). The median time to progression is 1.6 months. The 12-month PFS was 7% (95% confidence interval [CI], 0.01, 0.20) and OS was 79% (95% CI, 0.59, 0.90). The median survival time is 23.2 months (Figure 2).

Figure 2.

Progression-free and overall survival of patients with relapsed Hodgkin lymphoma receiving single-agent galiximab.

The number of cycles of protocol therapy given ranged from 1 to 15. Fifteen patients (51.7%) received induction therapy, but either progressed or were removed from study at the time of first evaluation prior to the start of extended induction (week 8). One patient received two cycles, five patients received three cycles and four patients were treated with six or more cycles of therapy. As mentioned above, one patient with stable disease remained on treatment nearly 2 years after starting galiximab.

Safety and tolerability

Galiximab was extremely well tolerated in this group of heavily pretreated patients, with few grade 3 or 4 toxicities (Table III). There were no study-related deaths and no one was removed from protocol treatment due to toxicity. The most common non-hematologic adverse events included fatigue, hypoalbuminemia, and hypophosphatemia.

Table III.

Toxicity of galiximab in patients with relapsed/refractory Hodgkin lymphoma^*.

Toxicity	All grades, n (%)	Grade 3–4, n (%)
Hematologic toxicity
Neutropenia	2 (7%)	0
Anemia	6 (21%)	4 (14%)
Thrombocytopenia	5 (17%)	3 (10%)
Non-hematologic toxicity
Fatigue	4 (14%)	1 (3%)
Pruritus/itching	2 (7%)	0
Infection	3 (10%)	2 (7%)
Increased ALT/AST	4 (14%)	2 (7%)
Hypoalbuminemia	3 (10%)	0
Increased alkaline phosphatase	2 (7%)	1 (3%)
Increased GGT	2 (7%)	0
Hyperglycemia	2 (7%)	0
Hypophosphatemia	3 (10%)	3 (10%)
Motor neuropathy	2 (7%)	0

ALT, alanine transaminase; AST, aspartate transaminase; GGT, glutamyl transpeptidase.

^*All toxicities occurring in more than 5% of patients are listed irrespective of attribution.

Correlative studies

Patients with baseline and first restaging PET/computed tomography (CT) scans with available SUV data are shown in Figure 3. Of these, 10 progressed at first restaging, and three progressed shortly thereafter. The median % SUV change was 24.65 (range, − 61.4 to 254.5). Given the low overall activity of galiximab, further correlation with clinical outcomes was not possible. None of the three responding patients (defined by CT criteria) had PET scans done to specification.

Figure 3.

Percent change in SUV_max at time of first restaging (8 weeks) in patients with relapsed Hodgkin lymphoma receiving single-agent galiximab.

Discussion

The past few decades have witnessed strong gains in targeted therapy for many malignancies. In HL, until recently, there have been few targeted agents in development. Many patients with relapsed and refractory HL are heavily pretreated, and those who relapse despite autologous stem cell transplant are either poorly tolerant of multiple rounds of additional cyotoxic therapy or display clear signs of chemoresistance. Galiximab, a primatized monoclonal antibody against CD80, is a rational agent to test in this setting and has an excellent safety profile. CD80 is normally present on activated B-cells, antigen-presenting cells and T-cells. It functions to regulate T-cell proliferation and T-cell effector functions via interaction with CD28 [7,15]. CD80 is nearly uniformly expressed in Reed–Sternberg cells, and anti-CD80 strategies have antipro-liferative effects in several pre-clinical models [4,5,15]. Specifically in HL models, the use of anti-CD80 immunotoxin leads to apoptosis and decreased cell growth [6,16].

CALGB 50602 (Alliance) was a multicenter phase II trial to determine the efficacy and safety of single-agent galiximab in patients with relapsed HL. The study accrued 29 eligible patients in 6 months, underscoring the demand for new agents in this setting. Of 29 patients receiving at least one dose of galiximab, there were no overt negative safety signals. Most adverse effects were mild in intensity, transient in nature and infusion-related. Unfortunately, the single-agent overall response rate was only 10.3%, with one complete and two partial responses; an additional nine patients had stable disease. The median PFS was 1.6 months, and only one patient has not yet progressed. Of note, the median survival time is almost 2 years, underscoring that some patients have a more indolent course despite non-response.

An exploratory aspect of the trial was to determine the predictive value of a mid-treatment FDG-PET following a biologic and non-cytotoxic agent. Our trial sought to determine whether or not changes in metabolic activity after induction treatment (8 weeks of therapy) would correlate with either response or PFS. Among 21 patients with paired pre- and post-treatment FDG-PET scans, there was a median SUV_max change of + 24.65%. In most patients, the increased uptake correlated with radiographic progression. However, in six patients, an SUV_max decline occurred despite radiographic progression or stable disease. Unfortunately, one-third of patients, including the three responding patients and the patient who remained on galiximab in excess of 2 years, did not have paired PET images available. In addition, the PET scan may have been done too early for a monoclonal antibody to have an eThect on SUV. Others have noted that functional imaging following targeted agents has mixed correlation with radiologic changes and may be related to the impact of the pathway being inhibited and overall cellular metabolism [17,18]. Thus, the metabolic changes in our trial are of interest, but these controversies as well as the low overall activity of galiximab precluded planned analyses of predictive capability.

Recently, the antibody–drug conjugate brentuximab vedotin has shown striking results in similar populations of relapsed and refractory HL, whereas predecessor “ naked” monoclonal antibodies against CD30 had disappointing results. Several anti-CD30 targeting agents have been tested. MDX-060 and the structurally related compound, MDX-1401, had limited activity [19,20]. Similarly, SGN-30 was assessed in both phase I and phase II studies, with no objective responses found in 59 aggregate patients [21,22]. In sharp contrast, conjugating the first-generation compound SGN-30 to an antitubulin agent, monomethyl auristatin E, dramatically changed the clinical activity profile of anti-CD30 targeting [23,24]. This antibody–drug conjugate, SGN-35 (brentuximab vedotin), achieved some degree of tumor response in 86% of patients and objective responses in 50% of patients treated at the maximum tolerated dose, thus leading to its Food and Drug Administration (FDA) approval for relapsed HL. A phase II trial of 102 patients with relapsed/ refractory HL subsequently showed a complete and overall response rate of 34% and 75%, respectively, to single-agent brentuximab vedotin, leading to FDA approval in late 2011. The consistent and extensive CD80 expression on HL raises the possibility that this could be an appropriate antigen for further targeting with antibody–drug conjugates despite the low activity of galiximab.

In summary, despite a promising rationale, there was minimal single-agent activity of galiximab in patients with relapsed/refractory HL. However, galiximab was well tolerated, and a substantial number of patients had stable disease. Further development of galiximab in HL is unlikely to proceed due to minimal activity and also due to production cessation by the pharmaceutical company. However, efforts should continue to develop non-cytotoxic and targeted agents for HL, not only to address multiply relapsed disease, but ultimately to replace cytotoxic agents associated with both short- and long-term toxicities in a typically young patient population.

Appendix

The following institutions participated in this study:

• Christiana Care Health Services, Inc. CCOP, Wilmington, DE; Stephen Grubbs, MD, supported by CA45418

• Georgetown University Medical Center, Washington, DC; Minetta C. Liu, MD, supported by CA77597

• Mount Sinai School of Medicine, New York, NY; Lewis R. Silverman, MD, supported by CA04457

• Southeast Cancer Control Consortium, Inc. CCOP, Goldsboro, NC; James N. Atkins, MD, supported by CA45808

• University of Chicago, Chicago, IL; Hedy L. Kindler, MD, supported by CA41287

• University of North Carolina at Chapel Hill, Chapel Hill, NC; Thomas C. Shea, MD, supported by CA47559

• Wake Forest University School of Medicine, Winston-Salem, NC; David D. Hurd, MD, supported by CA03927

• Washington University School of Medicine, St. Louis, MO; Nancy Bartlett, MD, supported by CA77440

• Western Pennsylvania Cancer Institute, Pittsburgh, PA; John Lister, MD

• Weill Medical College of Cornell University, New York, NY; John Leonard, MD, supported by CA07968